UMR: a multi-round algorithm for scheduling divisible workloads

In this paper we present an algorithm for scheduling parallel divisible workload applications. Our algorithm uses multiple rounds to overlap communication and computation between a master and several workers. We use "uniform" rounds, i.e. a fixed amount of work is sent out to all workers at each round. This restriction makes it possible to compute an approximately optimal number of rounds, which was not possible for previously proposed algorithms. In addition, we use more realistic platform models than those used in previous works. We provide an analysis of our algorithm both for homogeneous and heterogeneous platforms and present simulation results.